Re: [PATCH 01/20] [v2] x86, intel: Introduce macros for Intel family numbers

From: Ingo Molnar
Date: Wed Jun 08 2016 - 07:02:13 EST



* Dave Hansen <dave@xxxxxxxx> wrote:

>
> Changes from v1:
> * added acks from a few folks
> * Took the redundant "MODEL_" out of the macro names (Suggested
> by Borislav Petkov and acked by others)
>
> From: Dave Hansen <dave.hansen@xxxxxxxxxxxxxxx>
>
> If you are cc'd on this code, please check _your_ code vs. the
> model list in "intel-family.h". Please make sure you have all
> the models listed that you intend to.
>
> Also, rather than trickling these in via all the various
> maintainers, should these just get pulled in to the x86 tree in
> one go?
>
> Problem:
>
> We have a boatload of open-coded family-6 model numbers. Half of
> them have these model numbers in hex and the other half in
> decimal. This makes grepping for them tons of fun, if you were
> to try.
>
> Solution:
>
> Consolidate all the magic numbers. Put all the definitions in
> one header.
>
> The names here are closely derived from the comments describing
> the models from arch/x86/events/intel/core.c. We could easily
> make them shorter by doing things like s/SANDYBRIDGE/SNB/, but
> they seemed fine even with the longer versions to me.
>
> Do not take any of these names too literally, like "DESKTOP"
> or "MOBILE". These are all colloquial names and not precise
> descriptions of everywhere a given model will show up.
>
> These have all been compile-tested. I also made a stab at
> dumping .o files and looking for unexpected deltas when I was
> just replacing magic numbers with equivalent macros.

So I've picked up this series and restructured it: I've created a single patch
that creates intel-family.h and have put it into x86/urgent. This eliminated
dependencies and allowed some of the patches to be queued in their natural trees,
in particular the 7 perf patches.

Thanks,

Ingo